Key Performance Indicators What does it mean for Hospital Authority?

Transcription

1 Key Performance Indicators What does it mean for Hospital Authority? Dr Koon-hung LEE Chief Manager (Cluster Performance) HA Head Office, Cluster Service Division

2 Healthcare Expenditure in Hong Kong and Around the World % Health Service Expenditure as Percentage of GDP (2009) 5.3% = 2.3% + 3.0% GDP Private Public United States Netherlands Source: (1) World Bank (2010 data) (2) Census and Statistic Dept (HK) Switzerland Canada Austria Belgium United Kingdom Japan Finland Australia South Korea Hong Kong Mainland China Singapore 4.0

3 Hong Kong Healthcare System Dual System Private Self-financed by patients Public Highly subsidized by govt 3.0% GDP 2.3% GDP 12% inpatients 72% outpatients (1) 88% inpatients (1) (2) 28% outpatients Public Health (2) (2) (2) Source: (1) GDP: 2010 (Census and Statistics Dept) (2) Inpatient (secondary & tertiary care) : Public-private share by in-patient bed day occupied in 2010 from HA and Dept of Health (3) Outpatient (primary care) : Thematic Household Survey Report No. 45, Census and Statistics Dept (data collected during Nov Feb 2010) 3

4 Healthcare Structure in Hong Kong Others 1% Private Doctors 56.1% Primary Care (distribution of consultations) Chinese Medicine Practitioners 15.2% Public Doctors 27.7% Secondary and Tertiary Care Hospital Authority 88% Private Hospitals/ Doctors 12% Source: (1) Thematic Household Survey Report No.45, Census and Statistics Dept, December 2010 (2) Public/private share by in-patient bed day in occupied in 2010 (HA & Dept of Health) 4

5 Establishment of Hospital Authority (HA) Established in 1990 under the HA Ordinance A statutory body tasked to manage all public hospitals and institutions

6 Funding 2011/12 Govt subvention (92%) Fee income Other income Manpower: > 60,000 staff Funding Input Other recurrent expenditure: e.g. drugs, medical supplies, maintenance, etc Recurrent expenditure: About US$ 5 Bn Size of Operation 41 public hospitals and institutions (~ 27,000 beds) 48 Specialist Out-Patient Clinics 74 General Out-patient Clinics Annual Service Output Inpatient and day patient discharges: 1.4M Accident and emergency visits: 2.3M Specialist out-patient attendances: 8.7M General out-patient attendances : 5.2M 6 Source: The 2011/12 Budget of HKSAR Government Head 140

7 Services delivered by 7 Clusters New Territories West New Territories West New Territories East Kowloon East Hong Kong East Hong Kong West Kowloon West Kowloon West Kowloon Kowloon Central Central Kowloon East Hong Kong East Hong Kong West

8 Distribution of population, budget allocation and service Distribution of population, output budget by clusters allocation and projected output by clusters for 09/10 New Territories West Cluster New Territories East Cluster 14.9% 18.2% 0% 5% 10% 15% 20% 25% 30% 0% 5% 10% 15% 20% 25% 30% Kowloon West Cluster Kowloon Central Cluster Kowloon East Cluster 26.7% 7.0% 13.7% 0% 5% 10% 15% 20% 25% 30% 0% 5% 10% 15% 20% 25% 30% 0% 5% 10% 15% 20% 25% 30% Geographical population Acute Geographical Inpatient WEs population Extended care 7.7% Hong Kong West Cluster 11.8% Hong Kong East Cluster Previous years Ambulatory and Community care Budget 0% 5% 10% 15% 20% 25% 30% 0% 5% 10% 15% 20% 25% 30% 8

9 Performance Monitoring Framework HA Board (Administrative & Operation Meeting) Government (Permanent Secretary of Food and Health) HA Head Office Cluster Chief Executives

10 Performance Monitoring Framework Accountability to Government Permanent Secretary of Food & Health Controlling Officer s Report (COR) Hong Kong Government Budget Quarterly Progress Report to FHB Accountability to Hospital Authority Board of Governance Chief Executive s Monthly Progress Report on Key Performance Indicators A set of Key Performance Indicators (KPI) since 2008 Including some COR items, clinical service indicators, human resources & finance Accountability to Hospital Authority Head Office Clusters Key Performance Indicators Report since 2008 Progress report on Annual Plan Targets and Funded Programmes Cluster Management Meeting & Directors Meeting KPI reports

11 KPI HR Clinical Services Finance Budget Performance Operations Service Growth Quality Improvement Efficiency Manpower Resignation / Turnover Rate Sick Leave Injury on Duty 11

12 Service growth in response to population change & ageing effect Hospital beds, community nurses & day places Day/ inpatient discharges & patient days Ambulatory (clinics & allied health) attendances Community & outreach visits Efficiency in use of resources Bed occupancy rate Average length of stay Day/ same day surgery No. of inpatient episodes per bed Productivity in DRG weighted episodes Net asset value No. of KPIs Clinical 66 HR 15 Finance 15 Total 96 Improvement as a result of technology advancement or new service quality & access initiatives Waiting times: Accident & emergency Specialist OP new case Elective surgeries (Cataract/TURP) Diagnostic radiology Radiotherapy Breast/Colon/Nasopharyngeal cancers A&E standardized admission rate Unplanned readmission rate MRSA infection rate Stroke: % CT/MRI in 12 hrs/ %in Stroke units Hip fracture: % surgery in 2 days DM: % HBA1c < 7% HT: BP % <140/90 Cardiac: post MI % statin Mental: ALOS acute inpatient Human resources Manpower position Resignation / turnover rate Sick leave / injury on duties Annual leave balance Finance Income & expenditure statement Medical fee income, waiver, write-off Drug consumption and stock holder period Capital expenditure Debtor analysis 12

13 1,000,000 No. of inpatient discharge episodes 600,000 No. of day patient discharge episodes 980, , , , , , , ,000 HA overall 300, , ,000 HA overall 860, % Bed occupancy rate (%) 6,800,000 No. of SOP attendances 83.0% 82.5% 82.0% 6,600,000 6,400,000 6,200,000 HA overall 6,000,000 HA overall 81.5% 5,800, % ,600,

14 What is Hospital Authority s Performance? Areas with improvement Waiting time for cataract surgery, Cancer radiotherapy, MRI scans Stroke: CT/MRI in 12 hrs, reduce unplanned readmissions Increase % of ESRD receiving haemodialysis Injury on duty, sick leave Areas with significant deterioration Specialist clinic (SOP) new case waiting time, except Surgery & Ophthalmology Significant variation between clusters SOP waiting time: Gyn weeks; Surg weeks Radio-diagnostics: MRI days; CT days hip fracture: 67 94% surgery in 2 days Injury on duty absence days per 100 full time equivalent

15 40% % of ESRD patients receiving HD treatment 35% 30% 25% 20% 15% 10% 5% 0% HKEC HKWC KCC KEC KWC NTEC NTWC Overall HA 100% % of Episodes With Plain CT/MRI Brain within 12 Hours since A&E Registration 95% 90% 85% 80% 75% 70% 65% 60% HKEC HKWC KCC KEC KWC NTEC NTWC Overall HA

16 Performance Indicators! WHY?

17 WHO - Goals of Health Care System Achieving good health for the population Ensuring health services are responsive to the public Ensuring fair payment system

18 Health Care System Reforms Increasing accountability Cost effectiveness Sustainability Quality improvement Enhance hospital performance based on scientific evidence or best practice model Performance monitoring system esp. hospitals as they consume more than half of of most health budget Veillard J et al 2005

19 NHS Performance Assessment Framework Performance Ratings determines resources allocation

20 WHO Performance Assessment Tool for Quality Improvement in Hospitals (PATH)

21 Others projects on hospital performance assessment Australian Council on Healthcare Standards Joint Commission Accreditation of Health Care Organizations International Quality Indicator Project Organization for Economic Co-operation and Development Ontario Hospitals Association etc Groene O, et al 2008

22 Performance indicators If we can t measure, we can t manage Main attraction : promise visible and concrete proof of performance Objectivity : agreed rules of assessment across all organization, deriving knowledge independent of its creators Religious and missionary Performance indicators quality improvements? Is it evidence-based?

23 Source of Control Aims of performance indicators systems External Assurance / Accountability Internal Quality Improvement Internal Formative Supportive Continuous quality improvement Nature of expected actions Punitive Summative Internal evaluation External Accreditation External accountability Boland & Fowler 2000

24 External summative indicator systems External verification of quality improvement and central control May be political, commercial Renewal of legitimacy Information may be useful for political debate Organizational structure inscribed with auditable system Policy imposed by assuming the selection and structure of indicators related to goals, values and purposes

25 Formative quality improvement indicators Links to Total Quality Management (TQM) approach and Continuous Quality Improvement (CQI) paradigm Used as a focus for feedback and learning, leading to improvement Vehicle to align objectives of staff and the organization Stakeholders discuss and agree which indicators to include

26 Conceptual problems with performance indicators systems Performance indicators framework may displace existing informal, unrecorded modes of quality assurance May become a ritualistic system that hides genuine information Not capable of showing why particular results are obtained Difficult to inform policy and programme modifications

27 Technical problems with Indicators selection Key performance indicators Indicators only capture a fragment of health care Too few indicators will miss important aspects Too many indicators will be impractical and too costly to maintain Variation in definitions failure to compare like with like Difficult to operationalizing indicators Conflicting objectives, over-riding political importance and short political time-horizons

28 Performance indicators selection Data availability and reliability Tendency to measure what is already there instead of correlate with system goals and objectives Data accuracy : difference in results reflect quality of data rather than quality of care? Flawed data likely increase confrontation and reduce co-operation Data collection by goodwill vs manipulation, e.g. summative purpose

29 Performance indicators selection Data validity and confounding Validity means reflecting the right attribute, in this case of the healthcare system, rather than attributes of patients, or of other non-healthcare characteristics Only need to relate to factors that are under scrutiny Potential confounding factors: socio-economic variation, case mix, comorbidity, severity Difficult and costly to collect data attributable to confounding factors Adjustment for confounding Standardization Group analysis Predicted vs actual Multiple regression

30 Example: Re-admission Rate Unplanned emergency re-admission defined as an admission via Accident & Emergency (A&E) department to the same specialty in any HA hospital within 28 days of discharge from the index episode (IE). This is an improvement over the previous definition which, in principle, only counted readmissions via AED to the same hospital within 28 days of discharge; without taking into consideration of any inter hospital transfers, discharge specialty and the subsequent admission specialty of the readmission

31 Illustration Patient 1 Discharge MED(Acute) Hospital A MED(subacute) Index episode (Single hospital) Patient 2 Transfer Transfer Discharge Hospital A Hospital B Hospital A Index episode (Linked) MED(Acute) Rehab(subacute) MED(Acute) 31

32 Illustration Index Episode 1 Discharge MED(Acute) Hospital A MED(subacute) Assign to Hospital A - MED Index Episode 2 Hospital A MED(Acute) Hospital B Discharge Rehab(subacute) Assign to Hospital A - MED 32

33 Illustration Discharge Readmitted via A&E to Any HA Hospital Assigned Specialty Y (Index episode) 28 days This index episode entails an Unplanned Readmission Direct admitted to Specialty Y The unplanned readmission is counted against the hospital where the initial admission of the IE occurred. 33

34 Question: Has the IEs entailed URs in 28 days? No Yes URR = Yes Yes + No 34

35 Un-planned readmission rate (%) Specialty-based unplanned readmission rate Overall average Clusters

36 Example: Standardized Accident & Emergency Admission Rate Standardized for age, sex, triage category, ambulance in A good measurement of appropriateness? 35% 30% 25% Overall average 20% 15% Clusters

37 Performance indicators selection Robustness, sensitivity and specificity Small numbers Random variations Year-to-year variations in league tables may lead to unnecessary praise or sanction Different levels of sensitivity and specificity are needed for different purposes Summative accountability decisions requiring higher accurancy Prescriptive indicators are often used descriptively in starting of performance evaluation programmes

38 Perverse incentives and un-intended consequences Tunnel vision focus on phenomena quantified Myopia pursuit of short term targets Measure-focused Focus on enhancing the measurement rather than the objectives Misrepresentation Misinterpretation Gaming Organizational paralysis due to rigid performance evaluation Freeman T 2002

39 Performance indicators system Dimension of hospital performance assessed Number of individual/ groups of indicators Development methodology Scientific tools: reliability, validity, sensitivity, specificity Involvement of stakeholdes Participation : voluntary vs involuntary Number of participants Data collection Public disclosure Feedback mechanism & time Budget of performance monitoring system

40 HA internal resources allocation system and influence on performance measurement Early days Hospital-based management Historical and bed-based funding Activity-based funding Specialty-costing information Reward efficient hospitals Population-based funding No explicit correlation with performance measurement Pay-for-performance Case-mix adjusted funding formula Separate systems of Key Performance Indicators Pilot of Quality Incentive Programme linking financial incentives with KPI

41 Conclusion Two principle uses of performance indicators: external accountability and internal quality improvement Indicators can be prescriptive, descriptive or proscriptive Performance indicators are seductive as being objective May be inaccurate, misleading and cause negative unintended consequences, perverse incentives and gaming Those using indicators should caution with interpretation Audit or data system cannot replace the informal quality control system in day to day practice of medicine

42